View PDF Version - RePub - Erasmus Universiteit Rotterdam

Lekanne Deprez et al.
A
S
""'~"'
n
D2:1S?S
:;;~:'l'"
"'
c',.,
Figure 4 A, Regional localization of probes for lod on no[­
mal ,hrOnlosomes 4 and 22. The area in which the NF2 gene has
been mapped is indicated. B, Schematic repre-sentation of reciprocal
t(4j22),crearing 4p+ and 22q- marker chromosomes inmeningi.
oma MN32. The breakpoints (indicated with aHows) are shown
accon:ling to DNA probes for chromosomes 4 and 22.
rence of two meningiomas one could argue that the
patient is in fact suffering from NF2. However, this is
not very likely, considering the finding that the tumor
cells contained hVo aberrant chromosomes 22 whereas
control cells of the patient displayed a normal karyotype.
It is known that meningiomas may occur following
trauma (Preston-Martin et al. 1980, 1983). This
patient suffered from several head injuries, which may
explain the occurrence of two meningiomas in this case.
In MN32 both chromosomes 22 are involved in
translocations. One copy of chromosome 22 (22q - )
is involved in a balanced translocation with chromosome
4: t(4j22) (pI6jqll). Both products of the reciprocal
translocation are still present in the tumor. In this
case, inactivation of a meningioma tumor-suppressor
gene at or near the translocation breakpoint represents
a very likely hypothesis. If this mechanism of inactivation
is operative, we would expect that the exact localization
of the breakpoint corresponds to the localization
of the gene. Hybridization experiments lIsing
hybrid cell lines segregating the reciprocal products of
the t(4;22) were performed to map the breakpoints on
chromosome 4 and 22. The breakpoint on chromosomes
4 is located between 04S62 (4p16.1) and F5 .53
(4pI6.3) (fig. 3C and D). This area has been investigated
intensively because it contains the gene presumed
to be responsible for Huntington chorea (Bucan
et al. 1990), The distance between 04S62 and F5.53
is estimated to be 3 eM, on the basis of multilocus
linkage analysis(Cheng et al. 1989). On chromosome
22 the breakpoint was mapped between D22S1 and
D22S15 (fig. 3A and B). The distance between 022S1
and 022S15 is at most 1 eM, and the cumulative lod
score between these loci is 5,35 at a recombination
fraction of zero {Rouleau et aJ. 1989; Zhang et al.
1990}. Thislocalization agrees with the localization of
translocation breakpoints in six other meningiomas,
which were all mapped at 22qll (Casalone et al.
1987; Maltby et al. 1988; Rey et al. 1988). A schematic
representation of the reciprocal translocation
t(4;22) is indicated in figure 4B.
The other copy of chromosome 22 (marker 22q + )
is also involved in a translocation, leading to a dicentric
chromosome: 22 pter--+qll::1pll--+qter. The reciprocal
product of this dicentric chromosome was
not found, probably because it lacks a centromere. As
022S1 is probably still present in two copies in the
tumor DNA (fig. 2), it seems that the breakpoint in
this marker is distal to 022S1. Sequences distal to
D22$1 are presumably present in only one copy in
MN32 (fig. 2), and we showed that these sequences
were present on marker chromosome 4p + as a result
of the reciprocal t(4;22), Therefore it seems that the
translocation breakpoint in the dicentric 22q + chromosome
is also located between 022S1 and 022S15.
This rearranged chromosome could have lost the
tumor-suppressor gene together with the reciprocal
product of the translocation. If this is the case, we
would expect that the translocation in the 22q + marker
is closer to the centromere than is the breakpoint in
the 22q - chromosome. It is also possible that this
translocation, too, disrupts the tumor-suppressor
gene,
In MN32 the localization of the t(4;22) between
022S1 and 022S15 on chromosome 22 is identical to
the position of the translocation t(11 ;22) (q24;q12),
which is found in most cases of Ewing sarcoma and
of neuroepithelioma (McKeon et al. 1988; Turc-Carel
et al. 1988; Zhang et al. 1990). However, the balanced
translocation in Ewing sarcoma is reminiscent
of that observed in chronic myeloid leukemia and suggests
that the t(11;22) leads to the activation of a
proto-oncogene rather than to the inactivation of a
tumor-suppressor gene as is the case in meningioma.
Therefore we would expect that the gene involved in
meningioma and the one involved in either Ewing sarcoma
or neuroepithelioma are different.
So far, two earlier reports have suggested a localization
of the meningioma tumor-suppressor gene. Both
are in conflict with the localization suggested by our
experiments, Dumanski et al. (1987) describe a men-
116